Modern Econometric Techniques Applied To Three Essays In Spatial Economics
KeywordsKnowledge Production Function
Zhang, Hao Helen
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
EmbargoRelease after 10-Aug-2017
AbstractFor Chapter 1: This paper offers a meta-regression analysis of the controversial impact of EU structural funds on the growth of the recipient regions. It identifies the factors that explain the heterogeneity in the size of 323 estimates of their impact recorded in 17 econometric studies. Heterogeneity comes from the publication status, the period examined, controlling for endogeneity, from the presence of several regressors but not from differences in functional forms. For Chapter 2: Recent spatial econometric contributions call for theory-driven spatial models and W matrices capturing actual and time-varying interregional linkages. This paper answers this call by developing theoretically Griliches' well-known knowledge production function to add knowledge externalities to it. They capture how human and private capital originating from one region benefit the creation of innovation elsewhere. Furthermore, we measure interregional interaction based on the actual flows of patent creation-citation and of migration of the educated workers. They have the advantage of capturing clearly the direction of the knowledge transfers. Their presence in the theoretical model leads to a reduced-form spatial cross-regressive model which differentiates better the role of each type of externality - and displays a better goodness of fit - than the spatially lagged model where spillovers depend on geographical proximity only. Both models are estimated on spatial panel data covering the dynamics of innovation across US states over the 1986-1999 period. For Chapter 3: The Ricardian framework is increasingly used for the study of the impact of climate change on farmland values. While most of the Ricardian studies assume no interaction between the geographical units under study, the few that do rely on traditional proximity-based dependence. In this paper we argue that since the larger share of agricultural goods produced by a state is not for its own local market, including interregional trade in the Ricardian framework provides new perspectives, avoids a missing variable bias and prevents erroneous conclusions. Our new framework is applied to the system of the U.S. states over the four most recent censuses (1997-2012) and demonstrate that climate and socio-economic conditions experienced in a state's trade partners have a significant role on that state's local farmland values.
Degree ProgramGraduate College